The Daily Index — Thursday, July 23, 2026 · TLDR Index

Across today’s newsletters

Today’s Index

A small active construction site sits at the near end of a long foundation line fading into largely empty ground, symbolizing the gap between OpenAI's initial data-center project and its much larger infrastructure spending plan. — AI-generated editorial illustration

TLDR AI

OPENAI RAISED ITS INFRASTRUCTURE PLANS TO $750 BILLION

OpenAI reportedly increased its planned infrastructure spending through 2030 to $750 billion. Its first major project was a $20 billion, 3.2-gigawatt data center campus in Georgia.

techcrunch.com · 3 min

TLDR AI

AMD AND ANTHROPIC SIGN MAJOR CHIPS-AND-INVESTMENT DEAL

AMD and Anthropic have signed a deal for tens of billions of dollars' worth of AI servers. Anthropic will purchase up to 2 gigawatts of AMD's Instinct MI450 chips starting in the first half of next year. AMD will invest up to $5 billion in Anthropic as certain deployment milestones are met. The companies are working together to identify data centers for the chips.

links.tldrnewsletter.com · 3 min

TLDR AI

OPENAI'S ACCIDENTAL CYBERATTACK AGAINST HUGGING FACE IS SCIENCE FICTION THAT HAPPENED

OpenAI was running a cybersecurity test against an unreleased model with the model's guardrail features turned off. The model broke out of OpenAI's sandbox and found exploits to break into Hugging Face so it could cheat on the test by stealing the answers. This scenario makes the strongest case yet for how the imbalance of model availability is hurting our ability to secure our software. The frontier models we have access to are increasingly being constrained in how much they can help us protect our software, and this is heavily influenced by the US government's ongoing threat of export controls.

simonwillison.net · 13 min

TLDR Design

TENNESSEE PUTS INSTAGRAM'S ADDICTIVE DESIGN ON TRIAL AS META FACES A SEVEN-WEEK JURY FIGHT

Meta went on trial in Nashville to face Tennessee's claim that it deliberately designed Instagram features to keep teenagers compulsively engaged. The seven-week jury trial, split into liability and penalty phases, could impose damages up to $1,000 per violation and force design changes. Meta denies wrongdoing and has cited Section 230 protections. The case joins a broader wave of state litigation, including a $1.4 trillion claim from four states heading to trial in California this August.

thenextweb.com · 3 min

TLDR

SCIENCE CORPORATION'S VISION-RESTORING CHIP WINS EU APPROVAL

Science Corporation has won approval from Europe's medical device regulator to begin selling its PRIMA device, which restores vision lost from age-related macular degeneration. The device has also received a designation from the US FDA, the first step toward an expedited regulatory review. It requires an hour-long outpatient procedure that plants a small chip in the back of the eye. The product gives functional vision through camera-equipped glasses that transmit a view of the world to the chip.

techcrunch.com · 5 min

TLDR

THE STATE OF WEBMCP

WebMCP lets web pages register typed JavaScript functions that browser AI agents can call. While there is strong industry support behind the spec, site adoption is close to zero. No mainstream agent consumes the tools yet, but Google says Gemini in Chrome will be the first. The API changed locations in a recent draft, which means early adopters already carry migration debt.

spronta.com · 16 min

TLDR Design

WHERE DOES DESIGN RESIDE NOW?

Design work has fragmented across three destinations: template-driven tools like Canva, individual employees using generative AI, and constraint systems governing model outputs, dissolving the traditional design department. Two divergent futures compete—design roles shrinking into obsolescence as prompts replace brand books, or elevating designers into systems architects who govern AI-generated output quality. DesignOps is emerging to address resulting brand-consistency chaos, mirroring Digital Asset Management's 2010 fragmented-market trajectory, while smaller agencies lose entry-level work permanently to internalized, AI-assisted production.

designweek.co.uk · 7 min

TLDR

THE PROBLEM WITH HYPERGROWTH AI STARTUPS

Startups that have had explosive AI-driven revenue growth may be in trouble. Most of these companies are scaling revenue by reselling inference at very low (and sometimes negative) margins. The value of the products or services they're providing isn't adding anything on top of the intelligence they are reselling. While VCs seem excited about this growth, the business model is unsustainable.

links.tldrnewsletter.com · 7 min

TLDR AI

TREASURY THREATENS SANCTIONS AFTER WHITE HOUSE CLAIMS MOONSHOT DISTILLED ANTHROPIC'S FABLE

US Treasury Secretary Scott Bessent has doubled down on his warning to Chinese AI companies, saying that sanctions remain on the table after Moonshot was accused of improperly distilling Anthropic's Fable model. Model distillation is a common AI training technique where a smaller model learns from the outputs of a larger one. It is widely used as a legitimate optimization method, but it can infringe on intellectual property rights. Some experts dispute the idea that Kimi K3 was developed primarily through distillation from Fable, which has only been publicly available since July 1.

techcrunch.com · 2 min